o artigo discute alguns problemas conceituais envolvidos no tratamento antropológico da ideologia, e argumenta que a maior parte das dificuldades decorre de uma preocupação teórica com significados humanos essenciais
Título original
anthropology and the analysis of ideology asad, talal
o artigo discute alguns problemas conceituais envolvidos no tratamento antropológico da ideologia, e argumenta que a maior parte das dificuldades decorre de uma preocupação teórica com significados humanos essenciais
o artigo discute alguns problemas conceituais envolvidos no tratamento antropológico da ideologia, e argumenta que a maior parte das dificuldades decorre de uma preocupação teórica com significados humanos essenciais
Anthropology and the Analysis of Ideology
Talal Asad
Man, New Series, Vol. 14, No. 4, (Dec., 1979), pp. 607-627.
Stable URL
htp:/flinks.jstor-org/sicisici=(
125-1496%28197912%292%3A 14% 3A4%3COIT%3AAATAOI 3E2.0,CO%3B2-U
‘Man is currently published by Royal Anthropological Institute of Great Britain and Ireland,
Your use of the ISTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
hup:/www,jstororglabout/terms.hml. ISTOR’s Terms and Conditions of Use provides, in part, that unless you
have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and
you may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
hupulwww.jstor.org/journals/rai. html
ch copy of any part of'a JSTOR transmission must contain the same copyright notice that appears on the sereen or
printed page of such transmission,
ISTOR is an independent not-for-profit organization dedicated to creating and preserving a digital archive of
scholarly journals. For more information regarding JSTOR, please contact support @ jstor.org.
hupulwww jstor.org/
Wed Mar 29 14:08:57 2006ANTHROPOLOGY AND THE ANALYSIS OF
IDEOLOGY*
Tatat Asa
University of Hull
‘This lecture discusses some of the conceptual problems involved in anthropological
treatments of ideology, and argues that most of the dificultes arie from a theoretical
preoccupation withesenta human meanings—asembodied in theauthenticsocial categories,
actions and discourses of given cultures. This preoccupation, i is maintained, i shared by
anthropologists who are often thought of as being radically diferent from each other, and it
account forthe dificulies they have encountered in conceptualsing Social change. The ist
and longer part ofthe lecture explores these difficultiesin some writings by Bloch, Bourdllon,
Leach and Mary Douglas. The tendency to reduce anthropological problems about the nature
and consequence of particular publie discourses tothe philosophical problem ofthe origin of
‘seental human concepts x noted. The tendency to see authoritative meanings asthe a priori
totality which defines and reproduces the essential integrity of a given social order is ako
criticised, on the grounds that it lavesan important question unasked: namely, how particular
political and economic conditions maintain or undermine given forms of authoritative
discourse as sytem. The final pat of the lecture i devoted toa general critique ofthe vulgar
“Marxist theory ofthe social function of ideology, with particular reference to the kinds of
reductionism which that theory undertakes in it attempt to determine essential meanings,
and the kinds of question which it fils to consider adequately
1
In the early seventies when the question of anthropology and colonialism
‘was first being publicly argued out in this country, there wasn understandable
reaction on the part of many British anthropologists against the exaggerated.
role attributed, in some of the cruder criticisms, to anthropology as colonial
ideology. In the excitement of indignant response it was often forgotten that
the really interesting questions concerned the ideological conditions of
anthropology, and the implications of these conditions for its discourse, and
not the very occasionally direct but on the whole insignificant practical role
that British anthropologists had played in support of British imperial structures.
Instead ofenquiring into the effects of ideological conditions on anthropological
discourse the argument often degenerated into assertions about the personal
motives or politics of its producers. However, I do not want to address this
problem directly here, but instead to begin with a general puzzle: the modesty
of anthropologists regarding the ideological role of their discourses in the
determination of colonial structures does not seem to be matched by a
corresponding scepticism regarding the role of ideology generally in the
determination of social structures which are the objects of their discourse.
* Malinowski Lecture for 1979, given atthe London School of Economics on 6 Match,
Man (NS) 14. 607-27608 ‘TALAL ASAD.
‘Their science as discourse, so iti said, is not determined by social reality. And
yet the social reality of which that science speaks is typically nothing but
discourse. Itis this apparent paradox I should like to explore in what follows.
But first a word of warning. What I say about anthropology here is not
intended as an empirical generalisation of the work of all self-styled
anthropologists—or even of that of the majority. For a long time now
anthropology has ceased to constitute a coherent field of intellectual enquiry,
and its present unity is institutional and not theoretical. My comments relate
therefore to particular strands within social anthropology which seem to me
to hang together in significant ways, forming a recognisable pattern that has,
become an obstacle to further development, and one which it would be worth.
examining more closely. I repeat, this pattern does not define the unity of
anthropology because it relates toa number of assumptions and tendencies that
are neither exclusive to it, nor for that matter shared by al texts which would
be called anthropological.
1
Questions concerning concepts and society, classification and social structure,
rules and social behaviour have always been of central concern to social
anthropology. In recent years a fresh impetus has been given to such questions
deriving from the study of language. This new impetus has occasionally been
represented as a radical break, a New Anthropology, and lines of battle have
been drawn on what are alleged to be matters of profound importance for the
theory and method of social anthropology. However, it is very doubtful
whether the fresh developments represent a totally new departure, or whether
the lines of battle define positions which areas different as they have sometimes
been claimed to be.
In fact, many of the basic assumptions and concerns of more recent writers
influenced by the study of language can be traced back to Malinowski, not
only of those writers who have been happy to acknowledge the connexion,
‘but also of many who have not. Malinowski’ critics, ranging from linguists
such as J. R. Firth (1957) and Langendoen (1968) to the publicists for a New
Anthropology like Ardener (1971), Henson (1974) and Crick (1976), have
largely seen in Malinowski a failed language-theorist. One is often given to
understand that despite his commendable emphasis on the importance of
learning native languages for fieldwork, Malinowski had no real understanding,
of advanced language theory. He is dismissed by anthropologists like Sahlins
(1976) for his crass utilitarianism, or noted briefly by linguists like Lyons
(1968) as someone who contributed that quaint but not entirely valueless
notion, ‘phatic communion’, to semantic theory. That Malinowski’ texts on
language also contain an anthropological theory of culture (which is not, by
the way, to be confused with his theory of basic and derived needs), has gone
generally unnoticed. Not only in his explicitly linguistic texts, such as “The
problem of meaning in primitive languages’ and Volume 2 of Coral gardens,
‘but also in some of his other writings such as his famous essay on myth, there
is present a notion of culture as an a priori totality of authentic meanings toTTALAL ASAD 609
which action and discourse must be related if they are to be properly
understood and their integrity explained." Itis this notion, ast finds expression
in the writings of more recent anthropologists, that I shall be questioni
But I do not want to give the impression—especially on this occasion—that
1 am adding one more critical voice to the long procession of Malinowski’s
detractors. Malinowski was not just a splendid fieldworker, and an amusing
polemicist rather than a good theorist, as the received version in social
anthropology hasit. His untidy but extremely interesting writings on language
do not only reveal a problematical conception of culture, they also show signs
of attempts to break away from that conception and to search for new ways of
representing and understanding discourse within the context of social life?
‘My concern here however is not with Malinowskis texts—but with those of
anthropologists who are still very much alive.
‘An emphasis on ‘meaning’ is to be found, whether implicit or explicit, in
much recent anthropological writing. A preoccupation with identifying and
constituting a priori structures of human meaning is shared by a range of
writings that are otherwise very different—those dealing with cultural
categories, symbolic representations, codes and communication, image
‘management, rational transactions, ete. And the preoccupation is present both
in the rationalist perspectives of those concerned to assert the universality and
priority of cultural clasification systems, and in the empiricist perspectives of
those concerned with what they take to be the ultimate datum of flesh-and-
bone individuals, interacting intelligently within the real world. This interest
in meaning is not itself new, although there is perhaps a new self-consciousness
about that interest. Anthropological texts with titles like Implicit meanings,
Rules and meanings, Transaction and meaning, Explorations in language and
‘meaning, “The management of meaning’, ‘Form and meaning of magical acts’,
“The politics of meaning’, seem to be somewhat more evident today than they
were a generation ago—and from authors who might be described as
‘empiricists or as rationalists, and sometimes even as rationalist and empiricist
at one and the same time. It is worth noting that Leach (1976) has recently
written about the complementary character of two major perspectives in social
anthropology, the rationalist and the empiricist, which he regards not as
opposed but as eminently compatible.> However it is not my intention to
stiggest reassuringly, as Leach appears to want to do, that rationalist and
empiricist anthropologists are together producing a sum of understanding in
which what is lacking on the one side is made up by the achievement of the
other. On the contrary, I want to try and indicate some things they have in
common, and to suggest why and to what extent these things are a source of
present theoretical weakness.
‘One aspect of this weakness can be seen in what is often admitted to be the
repeated failure of social anthropologists to produce a viable theory of social
‘change. The reason for this may not be, as itis sometimes proposed, that ‘real’
factors of social change are many and complex, which is a practical difficulty,
but rather that the way the object of change is itself conceptualised makes the
possibility of such a theory difficult if not impossible.
In the writings of both empiricist and rationalist anthropologists, then,610 TALAL ASAD
human meanings are seen almost everywhere. More precisely, the basic social
“object (called society, social structure or social order) which is presented in the
discourse of such anthropologists is constructed out of essential human
‘meanings. Society as the realm of convention is generally opposed to nature as
the realm of necessity. In extreme cases, as in Sahlins's most recent dialogue
with Marxism, there is a tendeney for even nature to be represented as the
product of human action and cognition, as inert raw matter ordered by
constitutive human meanings. In a striking passage, which must surely stand
as one of the most lyrical expressions of the humanist project in modern
anthropology, Sahlins in the conclusion to Culture and practical reason writes:
nature ast exists in itl only the raw material provided by the hand of God, waiting
tobe given memingfl shape and content by the mind of man Is asthe lock of marble to
the finished statue; and ofcourse the genius of the sculptor—in the sme way a the technical
development of eultire—consis of exploiting the lines of defraction within the material to
hs own ends. That matble is refractory: there ate certain things one cannot do with it—such
are the facts of nature and the action of selection. But its the sculptor who decides whether
the statue isto be an equestrian knight contemplating his victories [...] ora seated Moses
contemplating the sins of his people. And i tbe objected that iti the composition of the
‘marble which compels the form of the statue, it shoul not be Forgotten that this block of
‘marble was chosen from among all posible ones becatse the sculptor saw within tthe latent
image of his own project (1976: 20).
This bold celebration of man's creative powers, with God complacently
occupying the rank of unskilled labourer, may not be to the taste of all
anthropologists. However that may be, this passage from Sablins does raise
sharply the question of how the basic social object presented in anthropological
texts is constituted in those texts, and the question of what some of the
theoretical consequences might be of the fact that the essential defining
elements of that object are human meanings and human projects.
‘My concern is not with the question of whether or not the anthropologist
should accept uncritically and reproduce directly in his model the explanations
produced by the people he studies. The arguments between so-called symbolists
and literalist (sometimes represented as a confrontation between soft, nostalgic
liberals on the one hand, and clear-sighted, unsentimental modernists on the
other) are still, afterall, arguments about the criteria to be used for determining,
the sense of what people in other societies say and do. On this question one may
note in passing that the attempt to make a rigid distinction between symbolic
and literal meanings is coming to be seen as a distinction that raises more
problems than it solves.* Recently, Sperber (1975) and Skorupski (1976) have
in their different ways discussed the confusions created by the anthropological
addiction to notions of the symbolic. That their own treatments are not free
of further difficulties is something I cannot consider now. In any case, my point
here has to do not with the proper interpretation of words, gestures, and thins
My concern in the first place is with the nature of the basic social object
constituted within anthropological texts themselves. And I want to argue that
in anthropological texts this object is typically constituted with reference to aTALAL ASAD. or
notion of ideology whose social significance derives from its being the
expression of an a priori system of essential meanings—an ‘authentic culture’
I shall try to explore briefly this conceptual object constituted of human
‘meanings and to criticise it as 1 proceed—not by pointing to a brute reality
which every sensible person must immediately recognise, but by interrogating
the texts. In the course of this interrogation I shall try to take note of the very
different models that structure the sense of ‘meaning’—the aim of a
speaker/actor, the rules of linguistic (and stylistic) conventions, the content of
an experiencing, construing mind, and the process of conceptual unpacking
and re-presentation that occurs through objective discourse—because part of
my criticism will focus on the troubles that arise from the combination of
what are often incompatible models in the anthropological attempt to define
authentic cultures.
M1
‘That the central object of anthropological discourse is primarily constituted
in terms of human meanings has very recently been reaffirmed by Maurice
Bloch in his 1977 paper “The past and the present in the present’.
In this paper Bloch is concerned to specify reasons for the perennial difficulty
that anthropologists have had in formulating an adequate theory of social
change. Bloch points out that the concept of social structure in anthropology
refers essentially to an integrated totality of social classifications and meanings,
a system of social rules and roles which can, in one widely accepted
anthropological sense of the term, be called ritual. Now if this concept of social
structure is linked to the doctrine of the social origin of concepts (the social
determination of cognition) it becomes impossible to specify how social change
‘can occur. This is because, so Bloch argues, a system of meaningful categories,
of shared concepts that makes communication possible (a system which is none
‘other than social structure) cannot explain the creation of new concepts.$
Bloch believes that the Marxist theory of determining infrastructure and
determined superstructure is no answer to the problem either, because:
the infrastructure is seen as external othe concepts ofthe actor, [And] for itt be a source
of criticism ofthe social order it means that people must apprehend i in terms avaiable to
them and which are diferent from and incompatible with those of the dominant social
theory. This means terms not determined by it. Otherwise the infrastructure, however
contradictory to the dominant socal theory, s never transformed into action and js cartes
‘on in ts own sweet way, totally irtelevane tothe process of history (1977! 281).
Bloch’s own solution is to propose that the different conceptions and
perceptions required for an effective criticism of the existing social order must
be determined by that which is other than society—in other words by nature.®
Now Bloch’s emphasis on the importance of ideological argument (which
has its roots in some fertile suggestions made by Leach in Political systems of
highland Burma) is certainly a move in the right direction—and I will come
back to it later. But even in such a discourse ‘social structure’ is presented as a
total integrated system of shared subjective meanings and conventions. Even
the Marxist notion of infrastructure is declared to be irrelevant to the processes,on TALAL ASAD,
of history’ unless it can be reduced to the experiences and aims of individual
actors. The problem of social change is thus presented as being a problem of a
change of concepts, the concepts available to ‘real flesh-and-blood individuals’
that can at once define and transform the essence of society.
‘The main difficulty with Bloch’s argument is this: the proposition that
universally valid concepts (or new and better forms of understanding) are
‘essentially generated by man’s encounter with nature (which is to be thought
of as being a kind of ‘undeceiving’ object) and not by his encounter with
society (which is to be thought of as being a kind of ‘mystifying’ object) may
‘or may not be acceptable. I would maintain that itis not. But even if it were,
it would not explain how people, who are presumably all in direct contact
with the same nature, come to have very different concepts within the same
society, and how they are able to engage in ideological arguments about the
basic transformation of social conditions in which they all live.
In other words, epistemological questions about the ultimate origin or the
final guarantee of social concepts and forms of knowledge (so beloved of many
anthropologists) are really quite irrelevant to this kind of problem. They
cannot tell us anything about the reasons why different kinds of ideological
position come to be held in social life, or about the ideological force or
effectiveness of particular political arguments. (That is not to say as some social
theorists both here and in France have recently done, that epistemological
problems are in themselves completely valueless, but that it is unnecessary, and
‘worse than unnecessary, to drag in epistemological questions when discussing.
problems regarding the constitution of what anthropologists want to call
social structure, or society, and the part that systematic ideologies can play in
the reproduction or transformation of that object.) In this respect, Bloch’s
mistake consists in making the assumption, which is by no means unique to
him, that society—and so too social change—is essentially a matter of structures
‘of meaning, structures which are at once the collective forms of experience, and
the social pre-conditions of communication (culture as language); and that
therefore, social criticism is not merely sometimes a necessary but always a
sufficient pre-condition for social change.
‘Thus anthropologists have presented the basic social objects in their texts
(social structure’) in such a way that they inevitably propose for ideology an
essential and determinate function—so that ideology is not only written in as
the basic organising principle of social life, the integrated totality of shared
meanings which gives that society its unique identity (its culture), but also
changing ideologies are sid to be essential to basic transformations.
In a recent issue of Man (December 1978). there appears a perceptive
‘comment on Bloch’s paper by Bourdillon. In it he observes quite rightly that
the equation Bloch makes between ritual concepts (or religious languages) on
the one hand, and hierarchy and exploitation (or social structure) on the other
is far too simple and therefore unacceptable.’ However, Bourdillon has no
quarrel with the essentially ideological constitution of that anthropological
object called social structure, On the contrary, he insists that social structure
“isa way of thinking about society which enables us to bring together a diverse
set of relations into a manageable system, and which expresses the continuityTALAL ASAD on
in any process of social events’ (1978: 595). As against Bloch, Bourdillon
maintains that this integrated system of roles, rules and rituals is essential for
any form of orderly social life, and not just a source of mystification which can
and should be done away with. Bourdillon is here of course echoing the
argument by Douglas (1970) that the dissolution of structure, far from being,
the rational aim of revolutions is merely the empirical cause of rebellions. But
he does not ask himself, as he might well have done: is the system which is
supposed to be necessary for thinking and speaking (whether speculatively, or
with authoritative force) about social life to be identified with what is necessary
for living in it?
which is supposed to be repr in the anthropologist’s definitive text (an
interpretative, reflective discourse). This ambiguity is not concerned merely
with the question mentioned previously of the degree to which the
anthropologist may or must draw on the explanations provided by the people
studied—sometimes referred to in the anthropological literatures the problem
of the informant’s model versus the anthropologists model. The question
being raised here is really concerned with the relationship the other way
fet me first illustrate this point with a familiar example.
Vv
Some years after Political systems of highland Burma was first published,
Gellner (1958) wrote a paper drawing attention to the book as
perhaps the most lucid statement ofa certain kind of Idealism that I know, and teachers of
philosophy could profitably we a selection of his statements as 4 means of explaining to their
Students what such Idealism is about (1958: 202)
Leach in effect sees other Functionaliss 2 holding a kind of Platonism. Only the static is
properly knowable, itis merely approximated by the empirically real...) Leach’s own
varlant to this sa kind of Hegelianis: reality changes because tis in confit, the conflict is
2 conflict of embodied ideas, and the change and confit are knowable by means of concepts
that are themselves in confit in a parallel way (1988: 193)
Gellner's paper contained some usefull points of criticism, but I believe he
was mistaken in his philosophical reading of Leach’s text. Besides, the
proposition that Political systems of highland Burma represents a view that is
appropriately described as Hegelian cannot be sustained by a careful analysis of|
that text. But that isall by the way. The interesting thing was Leach’s responseou TALAL ASAD.
in his Introductory Note to the 1964 reprint. There, you may remember,
Leach wrote as follows:
Incidentally, in friendly comment, Profesor Gellner has written off my whole argument
2s one of idealist error’. Truth and error ae complicated matters but it scems to me that in
Suggesting indirectly tha the Kachins have a rather simple minded philosophy which
presumesa relationship berween ‘idea’ and realty" not very diffrent from that postulated by
Plato, tam noe arguing that Pato was correct. The errors of Pltonism are very common
‘errors which are shated not only by anthropologists but ako by the people. whom
nthropologiss tu.
‘This attribution by Leach of the Platonic doctrine of forms to the Kachins
is quite astonishing—not because the Kachins are too simple-minded to hold
it, but because the doctrine is esentially a metaphysical one, and there is no
evidence in the text of Kachin metaphysics in this sense, Whatever the ‘errors
of Platonism’ may be, they are not the errors of everyday political and
economic life (which is what the book deals with) but the errors of systematic
philosophical speculation.
Such is the awe in which philosophy is held by anthropologists, that Leach
felt constrained to reply to a philosophical charge in philosophical terms. And
yet there was no need for this whatever. Gellner had attributed a particular
philosophical theory of meaning to Leach,* and it was in this context that he
criticised him for assuming that one needed changing concepts to know
changing reality. But Leach’s text is not based on a specific philosophical
theory of meaning at all:? a basic argument of Political systems of highland
Burma (like that of Bloch's paper) is that there must be changing social concepts
for such a thing as social change to occur. Now such an argument could have
been countered—by pointing, not to a faulty philosophical theory about the
relation between “ideas” and ‘reality’ (as Gellner did), but to an ideological
conception of social change. More precisely, it could have been countered by
drawing attention to a very questionable theory of culture—the theory which
gives logical priority to the system of authentic meaning supposedly shared by
an ideologically-defined community, and independent of the political activity
and economic conditions of its members.
Political systems ofhighland Burmaisrightly regarded by many anthropologists
as a most important text—although its remarkable originality and its failure
have rarely been adequately appreciated. But note how arguments about
systematic concepts within anthropological discourse can be easily shifted into
another key—and thus stopped, by an appeal to the discourse of the natives
“out there’ which the anthropologist has witnessed and recorded. Since Gellner
did not respond, we can perhaps assume that he accepted Leach’s authority in
this argument. However that may be, on this occasion, as on others in which
anthropological discourse secks to re-present an authentic system of human
‘meanings (the enduring categories which define, from one historical moment
toanother, who ‘the Kachin’ are), no native was available to contest the system
being imputed—if not to him or her individually, then at least to the ‘society’
which was supposed to be authentically his or hers.
Here is one example, then, of the way in which the anthropologist’s
discursive object comes to be presented as a reproduction of the essential‘TALAL ASAD ors
discourse of a whole society. (Although, be it noted, it is not the Kachin but the
anthropologist who actually writes) The example may be unusual in substance
but perhaps because of that it serves all the better to raise the question of how
anthropological texts construct for a whole society, or even for a group within
it,a total, integrated semantic system, which defines for that society what its
essential identity is
Let me pursue this question with further reference to Leach's Political systems
of highland Burma.
‘The significance of Leach’s text derives not from the fact that it illustrates a
simple philosophical error (idealism) which should be replaced by a sounder
epistemology but from the fact that it is a rich and complex statement of a
particular anthropological problem and its proposed solution. The text is not
simply the reflection of an essential philosophy. It is the production of an
anthropological object. I cannot here discuss this work in the detail that it
deserves, but will draw on an aspect of it which is relevant to my theme.
The starting point of Leach’s study, of course, was the question of Kachin
identity. The problem was that the so-called hill peoples of the north-eastern
Burma frontier region were rather diverse in their culture, lived in contrasting
ecological settings, spoke a number of quite distinct (often mutually
unintelligible) dialects, and were organised in local communities which
apparently held to very different political principles—varying from the
comparatively rich, autocratic Shan princedoms to the often rather poor, and
relatively egalitarian Kachin gumlao village domains. Leach’s answer to this
basic problem was that Kachin identity was based on a common ‘ritual
language’, an ideological system whose primary organising categories were
politica.
Leach’s notion of a common ritual language enabled him to rationalise
apparent local divergencies in the Hills Area into conceptual moments of the
‘same’ social structure, and apparent historical changes into elements of the
transformational logic of that structure. This rationalist solution to an empirical
problem was thus in effect the construction of a system of human meanings
(a “grammar of ritual action’ Leach called it) which was then identified as the
essential language that defined the political economic integrity of the Kachins—
and in terms of which the Kachins must speak if they are to remain
authentically Kachin. In this way Leach’s text defines what can and what
cannot be ‘correctly’ said in the political discourse of the Kachins.
‘Of course Leach allows for and refers in his text to ideological argument
among the Kachins. But the arguments described revolve in general around
the abstract principles of hierarchy versus equality, and Leach relates the way
in which the authoritative Kachin ritual language is used to claim one or other
of these two abstract principles, which are at once the principles of prevailing
social conditions and the principles of meaningful discourse. The resulting
account of social change, as many critics have observed, is thus an account of
an eternal cycle—better described as a process of ideological self-reproduction.
The difficulty here is not that there cannot be argument and criticism of
social arrangements if there is only one language which is “determined by
society’. The difficulty isnot that Leach should have looked more to" economics616 TALAL ASAD
and politics’ (where allegedly Man isin direct contact with Nature) and less to
ritual and religion in order to identify the origins of Kachin concepts of
Reality. The difficulty resides in the very notion of a “grammar’ which is at
‘once the principle that defines the anthropologis’s object of discourse and also
the system of concepts which is held to integrate and define Kachin political
and economic life as a whole. This ideological definition of the Kachin system.
misses the question of whether there are not specific political economic
conditions which make certain rhetorical forms objectively possible, and
‘authoritative. For when Kachins become Shan, for example, the process
involved is not merely a matter of the mental or behavioural change of
subjects, but ofthe partial undermining of a given form of discourse, and of the
production and reaffirmation of another form, within the very different
arguing, from the fact that the basic social object it presents is constructed out
‘of an integrated system of shared meaningfil ideas’, and so from the fact that
a closed, definitive status is given to that system (otherwise it would not be
‘integrated’ or ‘shared’ from one generation to the next), and all this in the
attempt to reproduce an authentic culture. What makes that system of
‘meanings ‘authentic’ is this very re-presentability from the past!® (from an
original generation to its authorised successors; and from the moment of
‘grasping it in the field to the moment of embodying it in the text). Thus the
political discourse of particular societies (as opposed to their knowledge of
Reality) is assumed to be self-defining and selfreproducing. Because of this
assumption, recent commentators like Bloch cannot see any way out of the
Leachean impasse other than by reducing the problem of the authority of
political discourse (which defines certain meanings as essential) to the very
different problem of the epistemological foundation and growth of objective
knowledge (‘Man's experience of Nature, direct/indirect’). And in this way an
anthropological question is answered in philosophical terms. Neither Leach
nor his later critics make any attempt to explore the systematic social
connexions between historical forces and relations on the one hand, and the
characteristic forms of discourse sustained or undermined by them on the
other.
In case it should still be thought that I am merely concerned to criticise
structuralist authors for their idealism, let me remind you that Leach's Political
systems of highland Burma isin its own way as materialist and as actor-oriented
asare the more recent texts by self proclaimed materialists or transactionalists.!?
v
But perhaps this argument can be made more strongly if we turn to Douglas
in her most recent, transactionalist mood—I refer of course to the booklet
entitled Cultural bias.‘TALAL ASAD or
Praising Cicourel for his attack on sociology and ethno-science, Douglas
comments:
Everything he writes about our collegues strikes this anthropologist 2s good clean fn, but
ieina pity that he never write anything about English social anthropology at all For we ate
hisnaturalallis, We abo believe that our work isto understand how meanings are generated,
caught and transformed, We aso astume that meanings are deeply embeded and context
bound. We ae ako stuck at che same fence that he has Baulked. Like him we cannot proceed
very far without incorporating rel lve cultures into our analysis For the cognitive activity
ofthe real live individual is largely devoted to building the culture, patching it here and
trimming it there, according to the exigencies ofthe day. In his very negotiating activity.
‘ach is forcing culture down the throat of his fellow men, When individuals transact, their
medium of exchange i in units of culture (1978: 5-6).
think we should not be distracted by these vivid images of threatened
authority—of horses that will not take the fence, and of prisoners who are
forcibly fed. What Douglas wants to know is whether culture is capable of
being radically altered, and her conclusion is that it is not. In order to establish
this, she attemptsa specification of the range of constraints on individual choice
and exchange, and this she does by the application of what she calls grid-group
analysis—an anthropological offipring of Bernstein’s well-known distinction
between elaborated and restricted codes. The resulting morphology of four
basic types of social context, and their supporting types of cosmology, form the
settings within which the ‘real live individual’ makes decisions, experiences his
or her environment, and transacts values with others
Douglas produces ethnographic examples from a wide variety of
anthropological studies of non-capitalist societies for each of the four main
cultural types. But illustrations for all the types are also provided from
industrial capitalist society, such that ‘individualism’, for example, can be
inscribed as a basic defining component of “middle class industrial culture’, as
well as that of highland New Guinea culture. And this is not without
significance, for she claims that one of the merits of her approach lies in its
ability to ‘cut across the clas structure’.
Douglas is quite clear about the futility of what she describes as class analysis.
Ever since [the eighteenth century] Europe's sl knowledge has considered social change
terms that ae based upon stratification, economic and polities! and upon occupational
tegorics For the anthropologis's projet, the stratified hierarchical perspective has not lent
‘lfc saying anything very Wseful bout the relation between culture and ideology on the
tone hand or between culture and form of socal organisation of the other. The present
‘exercise in understanding coxmologis is intended to cut a diferent kind of slice into socal
reality (1978254).
For Douglas, culture is represented as structured field of authentic meanings
‘on which individual experience, social interaction, and collective discourse are
all in different ways parasitic. The total culture, which mediates between
ideology and social organisation, is at once the collective precondition and the
long-term residue of meaningful choice and experience. And because social
facts are represented in terms of the shape and content of subjective experience,
rationalism and empiricism become, in her text, complementary modes of
accounting for the origin of those facts—a typical piece of reductionist
reasoning. Furthermore, since the particular pattern of meaningful individualors ‘TALAL ASAD
interaction, and the pattern of collective cosmologies, together define the
ultimate formation of social conditions in their four contextual variants, the
experiencing, transacting individual has really only one of two options: either
to adjust to a given integrated context (ike learning to use a language
correctly)—or to leave it altogether for a more congenial social context (or
language). And this is so because, in her own words, “Each position on the
[typological] chart is presented as an integral unit incorporating cosmology
and social experience as a single close-meshed structure’ (p. 41),
‘Thus for Douglas, the transformation of social structure is impossible, or
impossible to understand, because there is no social object that is specified
independently of a system of human meanings, and because such a system, like
4 given language, has the function of rendering the structure of cultural
experience and of political action isomorphic. As in Natural symbols, the
cultural and political pre-conditions for saying and doing things, as well as the
‘meaningful statements and actions produced in those conditions, are neatly
fused together. Nothing can be said or done with meaning ifit does not fit into
an a priori system, the ‘authentic’ culture which defines the essential social
being of the people concerned. The process of radical transformation is
described quite literally as ‘an emptying of meaning’—and quite rightly so if
the re-presentation of essential meanings is the mark of an authentic culture.
There is no space in Douglas's text for a concept of forms of social life which
have their own material conditions of existence, their own relations and
tendencies—that is to say, conditions, relations and tendencies which cannot be
reduced to the origins of human meaning, whether collective cognitive
categories or individual social activity. The concept of mode of production,
and the related concept of class structure, are of course precisely such concepts.
‘These concepts are not ‘slices of social reality’. They are not the “true objects”
of human experience. And they are not the ultimate origin or guarantee of
everyday categories or of languages or of ideologies. Such concepts are for
theorising the systematic historical aspects of social forces and relations by
which the material bases of collective life are produced—forces and relations
‘whose existence is distinct from that of individual meanings, intentions and
actions. I need not elaborate on this point here except to note in passing that
concepts of forces and of relations must be historicised, otherwise theorisation
in terms of modes of production is bound to become legalistic and/or
idealisti.? What I want to emphasise is that social life is not simply a matter
of systems of meaning (whether conventional or intentional), even ifit is true
that communication between human beings is necessarily present in every
domain of social activity—that social life is not identical with communication,
although communication is necessary to it. The logic of historical structures,
based on the forces and relations of production is quite different from the logic
of specific human intentions, of specific human languages, and of specific forms
of human understanding!
Not only is the logic of ideological structures different from the logic of
modes of production, but the former is also, in complex ways, dependent on
the latter." Put more concretely: the ‘individualism’ of highland New Guinea
societies cannot be assimilated, as Douglas tries to assimilate it, to the bourgeoisTALAL ASAD, 619
‘individualism’ of the middle classes in advanced capitalist society. The
parameters of the bourgeois ego are defined by an authoritative discourse
Which is rooted in material conditions profoundly different from those that
sustain the relevant discourses of highland New Guinea. For, to take a crucial
aspect of authoritative discourse: it is only in particular kinds of social and
material conditions that given forms of performative utterance can be
cffective.'® But for these utterances to be effective it is necessary that they be
understood and accepted in appropriate kinds of situation by appropriate kinds
of person. That is to say, given forms of effective performatives presuppose
certain types of ego, just as particular kinds of performative can be effective
only in the right material and social conditions. The difference between the
New Guinea Big Man and the bourgeois ego is therefore not merely one of
degree.
‘What can be said, in short, about all these anthropological ai which
nd, of course, once this
2a
vi
Perhaps this is the place to consider very briefly some popular misunder-
standings of Marx's famous formulation about the relation of ideology to the
material conditions of existence. In this whole area too many different kinds
of question are often confounded, by defenders and critics alike of that
formulation in anthropology. Let me try, however sketchily, to sort some of,
them out.
‘Take first ideology as systematic forms of socially constituted knowledge.
‘The Marxian proposition that there are specific material conditions for the
existence of specific ideologies in this sense does not necessarily imply that the
ideologies are simply effects or reflections of those conditions. Thus if we say
that the domains of systematic discourse within which specific knowledges are
produced, tested and communicated, are dependent on specific institutional
conditions and relations, this does not require any commitment to the view
that the knowledges reflect those institutional conditions—that there is an
isomorphic relation between the structure of knowledge and that of
institutional conditions. If we say that given social conditions sustain, and at
certain stages become obstacles to the development of scholarly understanding,
this does not imply that the concept of objective knowledges is nothing but
illusion. If we say that particular modes of systematic discourse can be and are
used for furthering particular class interests, this does not imply that all such.
modes of discourse are essentially nothing but an ‘expression’ of the positions
that are supposed to define those interests. So much may be familiar enough
and acceptable to many of you.620 TTALAL ASAD
But where the notion of ideology relates to notions of politics and social
funetion it becomes more problematical
Discourse which seeks to reflect on the nature of social conditions in a
systematic way can, in the process of being re-stated, contested, acted upon,
have some critical consequences for given conditions of social life. But ideology
assuch cannot be said a priori to have a universal, determinate function, because
what verbal discourses (as well as other modes of communication) signify or
do can only be determined by analysing the concrete social conditions in
which they are produced. It is no accident, for example, that attempts to
specify a determinate set of rules for defining performatives in English—let
alone in all anguages—have not succeeded.!° Beca
can develop a proper sleace ofthe symbolic, as tome anthropologists have
argued. It's: ‘terete eters ly, a theory which
will specify the universal pre-conditions, significances and effects of discourse.
This is, of course, in itself not a novel argument."? But it must be said that it
is one that stands opposed to the position not only of many anthropologists but
shot many Maton the quent nor Me™
Ttisall very well to say, as many Marxists and anthropologists do, that given
forms of social organisation (or given relations of production) always require
Gran logic for cutora) spac to manna snd repos tern Bo
Screg creme eben Romany moo eo
place such a formula is either tautological—as when ideology is said to define
relations of property so that what has to be maintained is identical with what
is supposed to do the maintaining; or it is reductionist—as when particular
utterances are said to have a predetermined impact on the ‘interpreting minds’
‘of those who uphold basic sets of social relationships, their sense being thereby
‘equated with their effect. What we can deduce, incidentally, from these
difficulties is that sometimes ideology is treated as social relation, and sometimes
as systematic utterance, and that in both its guises it is sensed as being mediated.
and structured, in an obscure way, by authorising discourses.
‘When the question arises of specifying, in concrete cases, what the crucial
ideological systems are, the anthropologist responds by postulating the presence
‘of an authentic system of meanings as the key to the discourse noted in the
field, and essentially reconstituted in the final text. It is this key that is used to
identify particular utterances as mere repetitions of the same discourse, and to
determine for them a mental effect as the crucial part of their meaning. In
other words, in order to establish the determinate function of a given
‘meaningful’ discourse, the anthropologist isolates what was said from its
thetorical context, and separates tendencies which might support given
conditions from those which might contribute to their undermining—taking
his theoretical separations for reproductions of the original. In this way, the
anthropologist’s text suppresses the tensions and ambiguities (conscious as wellTALAL ASAD. oar
as unconscious) that obtain within a given field of discourse in specific historical
conditions, and thus suppresses also the process by which motives, rhetorical
devices and forms of comprehension are constructed and reconstructed. Itis of
course precisely these ambiguities of discourse, and the elaboration which they
call for, that. make political argument possible. Yet even when the
anthropologist’s text presents two or more ‘competing ideological systems’
(official and unofficial, say), actual discourse is generally reduced to something,
celse—something that has a determinate social role which can be definitively
established in a neutral fashion by the analyst.
‘There is another (not unconnected) difficulty with the doctrine of the
‘maintenance function of ideology. When ideology is presented as the culturally
inherited lens of a given society by which external reality is filtered and
internalised for its members, or as the system of symbols by which their direct,
‘experience is rendered uniquely communicable, a well-known paradox is
created. For in doing so the anthropologist’s text claims for itself the ability to
represent that external reality directly, or to reproduce that inner experience
through very different symbols which are nevertheless assumed to be
appropriate—abilities which the exotic peoples studied necessarily lack. In
such an anthropological exercise, historically specific discourses are typically
reduced to the status of determinate parts of an integrated social mechanism,
and an epistemological paradigm which purports to define the problem of
‘objective knowledge is passed off as a sociological model for analysing
ideology.
It may be suggested here that the possibility of such a reduction is located in
the absence, in anthropology, of an adequate understanding of authoritative
discourse—ic. of materially founded discourse which seeks continually to pre-
tempt the space of radically opposed utterances and so to prevent them from
being uttered.2! For authoritative discourse, we should be careful to note,
authorises neither ‘Reality’ nor Experience’ but other discourse—texts, speech,
visual images, etc., which are being structured in terms of given (imposed)
concepts, and reproduced in terms of essential meanings. Even when action is,
authorised, it isas discourse that such action establishes its authority. The action
is read as being authorised, but the reading and the action are not identical—
that is why it is always logically possible to have an alternative reading.
‘The problem of understanding ideology is therefore wrongly formulated
when it is assumed to be a matter of predicting what ‘real’ or ‘experiential’
social forms are necessarily produced or reproduced by it. And this is so not
because forms of utterance never have systematic consequences (performatives
do, given that the relevant premisses are understood and accepted) but because
the effectiveness of such utterances is dependent on conventions which are
viable only within particular material conditions.
vil
‘One consequence of what I have been saying of course is that the vulgar-
Marxist view of ideology asa coherent system of false beliefS which maintainsoa TALAL ASAD
a total structure of exploitation and domination cannot really be sustained.
This is because, as I have already implied, such a view attributes at once too
much to ideology and also too little: too much because the maintenance and
continuity of a total social formation is supposed to depend on an integrated set
of concepts, and too little because the discourse in which ideologies are
articulated is identified as having an esential, univocal significance which
establishes at once its status as false belie, and as social determinant. This view
is not central to Marx’s own analysis of capitalism, but it does have a certain
‘currency among some Marxists.
It should not be imagined, however, that this view of ideology is peculiar to
vulgar-Marxists. On the contrary, like the conception of the integrated social
formation which many French Marxist anthropologists have been theorising
(and which I have criticised for some years) this view of ideology is a central
doctrine of functional anthropology.
Gellner’s discussion of baraka among the Mustim Berbers of Morocco??—
to take a well-known example—is very much in line with this particular
view of ideology. His analysis is concerned, in his words, with showing that
“There is here a crucial divergence between concept and reality, a divergence
which moreover is quite essential for the working of the social system’
(1970: 142). Gellner describes how a distinctive minority of Berbers, called
Saints, who act as mediators and arbitrators among the feuding tribes around
them are believed to possess divine quality, barakea, a quality which is thought
to be the origin of their authority and influence. But although the Saints are
believed to be selected by God for this office, so Gellner argues, the reality is
quite otherwise, because it is the tribesmen themselves who, by resorting to
these mediators, in effect select and accord them their authority—but without
knowing that this is what they do. In Gellner's memorable phrase, ‘What
appears to be vax dei isin reality vox populi’ (1970: 142). By which we learn
that certain Islamic religious doctrines are essentially mystified appearances of
political reality.
This whole style of anthropological analysis is based on what might be
called the Wizard of Or theory of ideology. Like Dorothy, the anthropologist
tears aside the veil of a seeming discourse to disclose the essential reality—an
ordinary-looking old man busily working a hand-machine. Even so, you will
remember the Tin Man really gets his heart, the Lion his courage and the
‘Straw Man his brain, so that belief, however absurd, is shown to have its social
function. But perhaps more important, you will also remember that afterall
the whole episode is Dorothy's dream—that the essential reality which is
revealed is itself a phase in the narrative of Dorothy's unconscious. By which,
of course, I do not want to suggest that ‘reality’ is insubstantial, but only that
the uncovering of “esential meanings’ is itself a production in discourse. In
‘other words, I want to remind you that Gellner’s text is the place where the
‘essential meaning of baraka is revealed/constructed, and so to pose the question
of the criteria by which that meaning is established. For the question that arises
here is this: Why do the surfaces of Berber religious concepts not reflect their
political meanings? Alternatively, why do the Berbers fail to see the
commonsense reality (the essential meaning) to which Gellner so easilyTALAL ASAD, 623
penetrates? Itis not much help to answer that Gellner is trained in philosophy
and in anthropology and that the Berbers are not, for that merely shifts the
problem from Geliner as the writing subject to anthropology as the field of
authoritative discourse reproducing other cultures, a discipline which defines
certain texts as competent. And bear in mind that the mistake the Berbers are
supposed to be continuously making is about the everyday commonsense
political conditions in which they live, and not about some final philosophical
Reality, Part of my point here is that there is an obvious difference between the
epistemological notion of mystification (implying direct v. indirect experience
of Reality) and the notion of mistakes or deception in everyday social and
political life—and that it is not very clear which notion Gellner is deploying
and why. But my main argument is that in either case the authority of Berber
religious ideology is left entirely unexplained. For baraka is not simply a
concept with a constructed (or disclosed’) meaning, but a part of authoritative
Berber religious discourse, and jt is no explanation of such discourse to say
simply that people believe in it And still less is it explained when we are told
that in essence the concept represents a delusion.”
Of course we can and should enquire into the political and economic
implications of religious discourse and argument for particular historical
conditions, and into the material preconditions which make such discourse
possible and authoritative. We can even attempt to establish particular instances
of political deception and error, as and when the evidence allows. But we need
not think that in doing this we are uncovering the true political or economic
essence of religious ideologies. Thus we do not have good grounds for arguing,
as many Western writers on modern Islamic reform movements have done,
for example, that the ‘real’ meaning of reformist discourse is not religious but
political, that authentic Islam is not reproduced in such discourse. I stress that
this is not a plea for respecting the true meaning of religious discourse—for
saying, as Evans-Pritchard said in Nuer religion, that the ‘real’ meaning of
religious concepts lies not in the external world of commonsense objects but
in the inner world of religious experience. Its not at all my intention to try
to rescue the ultimate integrity of personal experience. On the contrary. What
Tam arguing here is merely that the whole business of looking for and
reproducing the essential meanings of another society's discourse (its‘authentic
culture’) should be problematised far more drastically than it has been in social
anthropology—just as, indeed, it has already begun to be problematised outside
social anthropology.2® The search for essential meanings in anthropology
invariably results in the treatment of ideology in a reductionist fashion (either
by reducing ideology to economic political conditions, or by reducing
economic political conditions to ideology) and in confounding it with
philosophical issues. And itis this treatment of ideology that isso characteristic
of functional anthropology in its desire to reproduce definitively the authentic
cultures of other peoples. Instead of taking the production of ‘essential
‘meanings’ (in the form of authoritative discourse) in given historical societies
as the problem to be explained, anthropology takes the existence of essential
‘meanings (in the form of ‘authentic discourse’) asthe basic concept for defining
and explaining historical societies.624 ‘TALAL ASAD
vu
Let me try to state in a few words the general position which underlies
much of what I have been trying to say (a dangerous thing to do, but even so,
probably necessary). It is an old position, but one which bears re-stating given
the present self-consciousness about ideology and about meaning. However
much we might, as professional talkers and writers, wish toaffirm the profound
importance of systematic discourse, itis difficult to avoid the obvious, but by
no means trivial, conclusion that political and economic conditions have
developed and changed in ways that are rarely in accord with systematic
iscourse. Or let me put it another way: it is surely neither the power of social
criticism nor the relative strength of competing social ideologies within the
societies studied by social anthropologists (in Asia, in Africa and in Latin
America) which explains why and how they haye become basically
transformed, but the historical forces of world industrial capitalism and the
‘way these have impinged upon particular political and economic conditions.
Of course political arguments can be important (and especially if they help to
‘mobilise powerful social movements) but perhaps never in the way, nor to the
extent, that we flatter ourselves they are important. Given that this is so, the
‘main trouble with much colonial anthropology—and with much contempo-
rary anthropology too—has been not its ideological service in the cause of
imperialism, but its ideological conception of social structure and of culture.
"Ics this totality to which Malinowski cefers when he writes for example, hat ‘Since the
whole world of thingsto-beexpresed changes with the level ofeleure, wth geographic
socal and economic conditions, the eonsequene thatthe meaning of word must be sways
fathered, not from a pave contemplation ofthis word, but om an analy of is factions,
Stith efzenc othe given culture: Each prizutve or barbarous tebe as wells each ype of
SSvlation, has is world of meanings"-- (1923: 303). According to Malinowski, the
‘thnographer’ view of language ir uperor to that of the philologist, ecrse wnlie the lite,
Se crti har crc acct wo hs coca oaity and En ehttoe corey inert the
meanings which nar fo peopl’ act an uerances "What "meanings" lo ext in 3
fiven pect of native culture? by-meaning understands concept embodied inthe behaviour
tthe naive in their nteres, or in thete doctrine Thus the concep of api fore for
Jnstance, exis inthe very way they handle thie magi. But the problem of ascertaining
that, for imsance, th concep of magical force embodied in native behaviour and in thet
‘vole theoretical approach fo map: and then of ascertaining tht they certainly have no term
Toe this concee and ean only vicariously expres itis, in spite of ts negative quay, the
fei problem of ethnographic lingutics (19. 68)
“ Malinowskis expt preoceupation with lexeography and grammar has dive
azenton of many of his anthropological ries (specially those who approach his writings
from a Sausurian perspective) away ftom his attempss to deal with the theton
Janguage. Tas been let rgely to the work of exceptional trary critics, hike
to draw our attention, albeit very briefly, this side of Malinowskt
"Thus the two anthropological viewpoints which I have here summarised as “empiricist”
anu Trtonalst” are to be regarded 8 complementary rather than ight oF wrong! (Leach
i768).
7Soe the interesting paper by G. Lakoff and M. Johnson “Toward an experiential
ilosophy: the ease ftom literal metaphor. (mimeographed), Lingustes Department,
Uiprey of orn Brey. 975. sogephed, ee
Shit ecm a fist sight a stange problem because it dificult see why some of the
actors ts certain point inthe socal process ann say this system ino good all let ws take
2 ee Took atthe station and build upa new sytem, The retin why they cannot, within the
‘heoreial framework dacused, hes in the analysed notion of the socal determination of
aspects of
ike (1950),TALAL ASAD os
ee eee eee er eee
oa Sy on an er te
TD coments where man isin most direct contact with nature that we Bind universal
once ssh eet on ih ae ed
‘our cognitive categories’ (Bloch 1977: 285).
Soe ee 7 ag Dl vie on hem’ hngeee—rie
cat te are ces el on om
Febropec feb soe arent ote elon
CE ae ee ne ese
ream Peaking, och eve a enguny, tem f ropmo i 2
seubtecale rt tary otmeing Thos sha sence loge
Fee ea.
similar way’ (Gellner 1958: 189).
Bee ane 8 tevaing pari tory of aning cat wrong ft
iceman hs esta eens eae peer tel
sce geo opr es ey ars ty ier
ils ee game pred wr ce Se ote
tl aa PES ash a at ete
gl cs ote Reg ln a um fm
fern eng helps rtd re fecha rng et agin)
fe eck Se fe ec ee om a
han nest itaietsion encod epee anna
eran of ang dh one
Stutt dtal Sa@ oct arinht ones
-rson in the structural s} ‘in which he finds himself for the time being’ (pp. 10-11), he is
rehome ee en Joe
SNL Se timate sna tet partons oe ep
ee Tata ats ct coe or tamceninue tr ee
See eee
fines te prot ers ee ee
rect het nd ee penn Br nt ep och ely se
Se eee ae cae
Rofo ease tess wee cosets} ey
Si wen es a Pe a a
ee ee
SE TE of macs!
reproduction’ in Benjamin (1970).
WY Marvin Harris (1969) was not mistaken in his reading when he praised that book for its,
a Nari ene ete ren Ms rece
Se a eS carci of
cae beri opr sone mt of pretn he ci
spe rene epi ed toe cg ie em
eine ola be mg mere oe ek ees
amen
fo ee
arson ae, be eden! chemo ga fol
ee es een eee
Baw Ye ere On a cet ec ae ev ee
See ee ere
Fee ee cla cae ce genie pe pena eae
st ranting grec, ions sree, 2 rm ols sms
eee mo. Noo haters pte econ eh
Means iceertiet nes ait ac aac a ea
ean of egy 8 tae hat hs em the gy of
Sicha str hxc one tag eee
“heritage ce
Se ee A ee et ee
Sea Eee ee a aie: ae626 ‘TALAL ASAD
to ance, be proposes heterogeneity of dicunve principles (itnctive ‘gramman’)
contradictory printpes of dicouse which in industrial capitals societies are situated on the
rel of can Conte There b always dnger In tendencies fepesened by tha ste
(tren and ngs choogh i) tha inthe end even Mai wl be ef wath neg
Steal Saou and rel opi’ who profucr tsa decane, whe sore poise
sconomies se backted on One m blng no more tha abaractons or hora
conspt
re leshinov’s Masi andthe philsphy of angus, fie pblbed iy years ago sil
‘one of the most fruifal dicusions ofthis problem, and akhough st leaves many dificult
‘hess unanswered, it tevens a entity Yo the complex socal foundations of dscoune
Mfc srpmae thot wint ws wenten on ts sabe ody.
"SSRutin (ints), who fe coned the ter "perfomative’ and then abandoned the
concept In his nrg esa stay of An, Craton (197) hse aged thatthe
‘Smept of performative’ propel detned, shouldbe reed And forthe wey seo ht ed
{S'the Ln giving penal, tbe impowbity of cashing, rough, + shap
‘Spastion btn Seg the worl axl merely concving fin arr wy.
Teter he tempt sat ban vith Searle What ia sperch ac .96s)—reprned in
Gigli (1973), Se Coulthard io77), chap for sme of the problems involved in
scitying. sod clang perorsttves (or loctionary sca) im English A fr eter
itaguet dK. Rese we in Bauman and Shersr 1974: 48 that Pp Raver whe.
tiers thn anyone ee as explored the eng and westneas of this hotion for com
‘Bltralcomparsons hs ecclvely angel gta an ener aplciion of perforate
Stans ford on Eoglth,o eter langage ind clare) Te rk of Raver whch
Shelby Roses euimeoprapied pope hich Ihsve-unferuney enor been sble to ete,
i Sich an ssumpton & forays Peden inthe fllowing rears bythe edo ot chit
valuable collection, Explorations nthe elmogrphy of pecking’. the ethnography of speak
Tithe gp inthe nthtopologial record cated bythe meget by antbroplogesl ing
the soca eof language indy the lick of interest of ethnographersn patterns and functions
fopesking The portance ofthe ehnogeaphy of speaking to abroplogy cas far Seper
than thi however fora carefl fcw on speaking sr an instrument for he cond of soil ie
Brings tthe fore ihe meget ste of al races, gly deine by he onal
Sur of ee ae ly ted nema y ie na nd eed
itn fraying Cnn Sc 19748, yeni, Fs en
arf the vel oficial ection end communiction whlch get the histor
Sroaure of ole avo propo by Kaper (1976 15<10)—and hone ofthe things
int wich | am arguing In this coment, it may be worth svesing thatthe deincton
Seve the condition of pole economy and tent of uote dts ot pre bc
othe anthropolopeal dition Seen ral nd stl norms
"Wher example Gert (1973: 208-9).
‘9"Thus im his mowe rent work, Roland Barthes hus sbandone his ey stems at
ASAD, Talal Anthropological Texts and Ideological Problems An Analysis of Cohen On Arab Villages in Israel, Economy and Society, V. 4, N. 3, 1975, Pp. 251-282.